the Importance of Soil Bill Hunt, Ph.D., PE, D.WRE Associate - - PowerPoint PPT Presentation

www.bae.ncsu.edu/stormwater

Low Impact Development and the Importance of Soil

Bill Hunt, Ph.D., PE, D.WRE Associate Professor & Extension Specialist North Carolina State University

MN Clean Water Summit – UMN Arboretum, MN – 13Sep12

www.bae.ncsu.edu/stormwater

The Real Reason I’m Happy to Be Here

www.bae.ncsu.edu/stormwater

Outline

• Soil Importance on Defining the Target

Condition

• Underlying Soils & Infiltration by SCMs
• Utilizing Soil / Media to Sequester/ Mitigate

Pollutants

– Phosphorus, Temperature, + MediaDepths

• Minimizing Compaction Impacts to Maximize

Infiltration

www.bae.ncsu.edu/stormwater

Low Impact Development

• Reduce impervious surfaces
• Retain runoff on-site
• Promoting infiltration and

evapotranspiration

• Soils were recognized by the “Founding

Fathers”

www.bae.ncsu.edu/stormwater

Early on… LID “First Step”

• How to Lay Out your site?
• Where to location your practices?
• Is ideally based on…

www.bae.ncsu.edu/stormwat er

www.bae.ncsu.edu/stormwater

Your Soils!

Courtesy of Todd Houser, Cuyahoga S&W

www.bae.ncsu.edu/stormwater

e.g., Cuyahoga County Soils & LID

• Loamy Sand & Sand (5.8%)

– Infiltration, if well drained – Wet or dry BMP depending on drainage class

• Well Drained Soils (12.6%)

– Dry BMPs

• Moderately Well Drained (11.2%)

– 1.5 - 3 ft. November to May most years – Shallower – Wet – Deeper – Dry with Liner

• Somewhat – Very Poorly Drained (76.2%)

– 0.5 – 1 ft. November to June most years – Wet BMPS

• Slippage Prone (5.8%)

– Stay Away

• Urban Complexes (49.9%)

– Highly Variable

KSU

( /

Courtesy of Todd Houser, Cuyahoga S&W

www.bae.ncsu.edu/stormwater

Goals of Low Impact Development

• Reduce impervious surfaces
• Retain runoff on-site
• Promoting infiltration and

evapotranspiration

• Replicating pre-development hydrologic

conditions as closely as possible

• Davis, 2005

Amounts of Each – Depending upon Underlying Soil

www.bae.ncsu.edu/stormwater

Primary Goal of LID

Design each development site to protect,

• r restore, the natural hydrology of the

site so that the overall integrity of the watershed is protected. This is done by creating a “hydrologically” functional landscape.

www.bae.ncsu.edu/stormwater

What is the Target Condition?

• In North Carolina, Coweeta

– Coweeta is a US Forest Service Research Station

www.bae.ncsu.edu/stormwater

Establishing Target Condition

• Coweeta Hydrologic Laboratory

– Forested Mountain Watersheds – Longest Term Hydrologic Record (Runoff, Infiltration, and ET) in the World for a forest site

• Records used for this study were 37 and 50 years old

– Rain Shed Mountain Region

• Annual Precipitation in excess of 1500 mm (60 inches)
• Drawback: Coweeta Wetter than any major city in NC

– Data From Swift, LW, G.B. Cunningham, J.E. Douglass. 1987.

“Climatology and Hydrology.” Forest Hydrology and Ecology at

• Coweeta. Eds: W.T. Swank and D.A. Crossley. Springer Verlag.

New York, NY, pp 35-57.

www.bae.ncsu.edu/stormwater

Establishing Target Condition

Annual Hydrologic Fate Average Amount Percent of Total Precipitation Precipitation 1770 mm (70 in) 100% Runoff 80 mm (3 in) 5% Evapotranspiration 890 mm (35 in) 50% Infiltration 800 mm (31 in) 45% Shallow Interflow 770 mm (30 in) 44% Deep Seepage 30 mm (1 in) 2% All values rounded to the nearest 10 mm or 1 in. Infiltration is the sum of Shallow Interflow and Deep Seepage

www.bae.ncsu.edu/stormwater

Santee Experimental Forest (Coastal Plain South Carolina)

Evapo- transpiration Infiltration + Runoff Experimental Data (Amatya et al., 2006) 77% 23%

 Coastal plain region, undisturbed woods, sandy soils  30+ years of experimental data

www.bae.ncsu.edu/stormwater

Why the Difference?

• Sandier Soil Systems + flatter landscapes +

higher water table systems = MORE ET

• Underlying SOILS!
• Soils impact what your HYDROLOGIC TARGET

CONDITION is

Example of Target Condition Variation Landscape Detention: NC

North Carolina Grassed Cell

20 40 60 80 100 120 140 1 2 3 4 5 6 7 Rainfall (cm) Outflow Volume (m^3)

NC Grassed Cell Pavement Woods B (CN 55) Woods C (CN 70) Regression (CN 79)

www.bae.ncsu.edu/stormwat er

So, how well do SCMs “Convert” Runoff to Infiltration?

www.bae.ncsu.edu/stormwat er

Answer: It depends (in part) on Underlying Soil

www.bae.ncsu.edu/stormwat er

Improving Infiltration

www.bae.ncsu.edu/stormwat er

Water Balance

www.bae.ncsu.edu/stormwater

Brown & Hunt, JEE 2011

Reduced Performance in Clayey in- situ Soils

• Rocky Mount (sand): Upper Coastal Plain
• Greensboro (clay): Piedmont
• Graham: (N) loamy-clay & (S) sandy-loam

“We Bring Engineering to Life”

Site # Events Monitored # Events w/ Outflow Media Depth (m) IWS Depth (m) RM Grass 78 5 0.9 0.6 RM Mulched 78 4 0.9 0.6 Greensboro 1 63 18 1.2 0.6 Greensboro 2 63 40 1.2 No IWS Graham (N) 40 34 0.6 0.3 Graham (S) 40 22 0.9 0.6

Ritter Field Stormwater Wetland (River Bend, NC)

www.bae.ncsu.edu/stormwater

Runoff Reduction by Storm

1.0 10.0 100.0 1000.0 10000.0 1.0 10.0 100.0 1000.0 10000.0 Inflow (m3) Outflow (m3)

Line if Outflow Volume= Inflow Volume

www.bae.ncsu.edu/stormwater Lenhart and Hunt. JEE. 2011

Wetlands are not “supposed” to reduce runoff volumes this much!

• Why is this wetland so “good” at infiltrating?

Design Element Value Watershed Size 115 ac Wetland Size 0.34 ac Watershed Curve Number 55 Underlying Soil at Wetland Appling Fine Sand

www.bae.ncsu.edu/stormwater

Ritter Field Stormwater Wetland: Why is it so (relatively) small?

Design Element Value Watershed Size 115 ac Wetland Size 0.34 ac Watershed Curve Number 55 Underlying Soil at Wetland Appling Fine Sand

Determining Volume: Using NRCS Methods

• NRCS Curve Number Method

– SA = Volume (V) ÷ Ponding Depth Depth (D) – V = Runoff Depth (Q*) × Watershed Area (A) – Q = (P – 0.2 S)2 ÷ (P +0.8 S) P= Precipitation Depth; S = Initial Abstraction – S = 1000/CN - 10 CN = Curve Number

www.bae.ncsu.edu/stormwater

Key “Mid-term” Take Home Points

• Underlying Soils impact performance of a

Stormwater Control Measure

– E.g., SCMs over sandy soil will infiltrate (sometimes much) more than those over clays

• Underlying Soils have major factor in size of

practice (& therefore cost)

– Sandy Watersheds can have much smaller SCMs than Clayey Watersheds

www.bae.ncsu.edu/stormwater

Watershed Soils & SCM effectiveness

www.bae.ncsu.edu/stormwater

• Stormwater is a transport mechanism for

bacterial pollution

• Urbanization can lead to increased pathogen

loads to surface waters

– Pet waste, sewer overflow, sewer leakage

• Leads to 50,000 acres of Shellfish closures in

NC each year.

Bacteria Pollution

www.bae.ncsu.edu/stormwater

Charlotte Wet Pond Piedmont Clays

Waters ershed ed = 48.6 6 ha CN = 75

www.bae.ncsu.edu/stormwater

Wilmington Wet Pond Coastal Plain Sands

www.bae.ncsu.edu/stormwater

Charlotte (Clay) Wet Pond – E.Coli

www.bae.ncsu.edu/stormwater Hathaway and Hunt. JIDE. 2012

Sand Underlying Soil Pond – E.Coli

www.bae.ncsu.edu/stormwater Hathaway and Hunt. JIDE. 2012

Why?

• Pathogen Indicator Species “travel” (in part)
• n sediment
• Coarser sediment = better trapping efficiency

for TSS – And therefore for E.Coli

www.bae.ncsu.edu/stormwater Hathaway and Hunt. JIDE. 2012

Fill Soils/ Media & SCM Effectiveness

www.bae.ncsu.edu/stormwater

Greensboro Bioretention

www.bae.ncsu.edu/stormwater

TP Removal/Sequestration Greensboro

0.5 1 1.5 2 2.5 3 TN NO3-N TKN TP Mass (Kg)

In Out

www.bae.ncsu.edu/stormwater Hunt et al. JIDE. 2006

“We Bring Engineering to Life”

Chapel Hill Cell, C1

STP/WS = 0.14 Conventional Drainage

TP Removal/ Sequestration Chapel Hill

0.2 0.4 0.6 0.8 TN NO3-N TKN TP Mass (Kg)

In Out

www.bae.ncsu.edu/stormwater Hunt et al. JIDE. 2006

Results of Early Research

• Relationship between P-Index (Soil Test P) and

TP outflow load.

Greensboro Chapel Hill TP +240%

• 65%

P-Index 85-100 4-12

(Hunt 2003)

P-Index 50-100: High P-Index 0-25: Low

www.bae.ncsu.edu/stormwater Hunt et al. JIDE. 2006

Blame it on the Media…

Phosphorus Index (P- Index) is a measure of how much phosphorus is already in the soil media. Low P-Index: Can capture more phosphorus High P-Index: Soil is “saturated” with phosphorus

Very High: > 100 High: 50-100 Medium 25-50 Low: 0-25

www.bae.ncsu.edu/stormwater

Enter… the NC “Standard” Fill Media

• 85% Sand
• 10% Fines (Silt + Clay)
• 5% Organic Matter
• + Low P-Index

– 10 to 30

• Time to test it…

www.bae.ncsu.edu/stormwater

Mecklenburg Co. Hal Marshall Bioretention Cell (2004-2006)

Soil – 80% Mason Sand – 20% Fines + Compost – P-Index = 6 – 4 ft (1.2 m) Depth

www.bae.ncsu.edu/stormwater

TP - Charlotte, NC (2004-2006)

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1/1/04 7/1/04 12/30/04 6/30/05 12/29/05

[TP] in mg/L

Date

TP-In TP-Out

Concentration Red. = 31%; Load Reduct. ≈ 50%

Hunt et al., JEE. 2008

www.bae.ncsu.edu/stormwater

www.bae.ncsu.edu/stormwater

Effluent Concentration vs. P-Index

Site P-Index Depth (in) Outflow (mg/L) C-1 6 48 0.13 L-1 1-2 30 0.16 L-2 1-2 30 0.18 G-1 35-50 48 0.57 G-2 85-100 48 1.85

L O W

Take Home Message

• Your composition of Fill

Soil / Media Matters

• Be careful of what you

add or incorporate

www.bae.ncsu.edu/stormwater

Sometimes the Question isn’t “What type?,” it’s “How much?”

www.bae.ncsu.edu/stormwater

Depth of Fill Soil/ Media Matters!

www.bae.ncsu.edu/stormwater

Results: Oil and Grease

20 40 60 80 100 0.1 1 10 Infiltration Rate, cm/min Oil/Grease Removal Effciency, %

100% O/G removal. Type of Media has no impact.

Hsieh & Davis J. Environ. Engg. 2005

PAH Concentration (g/g dry)

5 10 15 20 25 Mid 30 to 41 cm Mid 20 to 30 cm Mid 10 to 20 cm Mid 0 to 10 cm In 41 to 51 cm In 30 to 41 cm In 20 to 30 cm In 10 to 20 cm In 0 to 10 cm Top Loose Gravel Top Crust (1-2 mm)

Accumulation - PAH

CP

Diblasi, Li, Davis & Ghosh, Environ. Sci Technol., 2009

Accumulated PAH are found in upper layers of sediment in bioretention

• facility. Very little PAH compounds have migrated deeper into the cell

media

1 2 3 4 5

5 10 15 20 50 100 150 200 250 300 Total Zinc (mg/kg) Depth (cm)

Original BSM Solid = Organic Empty = BSM

Depth and distance - Zinc

Zinc concentrations decrease with distance from the inlet and media depth.

Jones & Davis, in preparation

Nitrate with Depth

• 60
• 40
• 20

20 40 60 80 100 20 40 60 80 100 120 140 % Removal Bioretention Depth (cm)

• c. Nitrate

Different Pollutants – Different Fill Soil/ Media Depths

www.bae.ncsu.edu/stormwater Hunt et al. JEE. 2012

Pollutant Recommended Media Depth (ft) References TSS, PAH’s <1 ft

Li et al. (2008); DiBlasi et al. (2009)

Metals 1 ft

Li and Davis (2008); DiBlasi et al. (2009);

Phosphorus 1.5-2.0 ft

Hsieh et al. (2007); Hatt et al. (2009b);

Nitrogen 3 ft

Kim et al. (2003); Passeport et al. (2009)

www.bae.ncsu.edu/stormwater www.bae.ncsu.edu/stormwater

Reference: Bartholow, 1981; Responsive Mgmt, 2009; Gartner et al., 2002; Wegge et al., 2012

Criteria Temp.

Salmonid incipient lethal temperature

25 °C (77 °F)

NC trout upper avoidance

21 °C (70°F) 2008 in NC Over 92,700 anglers =

$174 million total

economic output

Coldwater Ecosystems Have Economic Value

2007 in AK

~$1 billion total

economic output supported 11% of regional jobs 2002 in MN

$150 million

in direct sales supported 3000+ jobs

5 10 15 20 25 30 May June July August September October

• Avg. Temperature (°C)

Runoff

• Conc. Inlet

Metal Inlet Effluent

4-8oC Increase

Jones & Hunt, JIDE. 2010

Jones & Hunt, JEE. 2009

Take Home Points: Fill Soils

• Fill Soils/ Media

Selection and Depth is hugely important in TP and TN performance

• Fill Soils/ Media

PRESENCE is equally important to “tame” temperatures

www.bae.ncsu.edu/stormwater

Improving Infiltration Capacity of Soils

• LID really is predicated
• n getting some runoff

(at least temporarily) into the ground

• We want to encourage

this.

www.bae.ncsu.edu/stormwater

Wilmington, NC (Anne McCrary) Permeable Concrete Study

• Loamy Sand Soil–

Coastal NC

• Undisturbed K ~

1-2 in/h

• Water table > 1 m

from surface

• Day Use

Recreation (40 ADT)

www.bae.ncsu.edu/stormwater

“We Bring Engineering to Life”

Permeable Concrete Wilmington NC

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 < 12 hour Rainfall (in) Runoff (in) Runoff from an Impermeable Lot Runoff from Permeable Concrete Lot LESS THAN 0.1 in/hr INFILTRATION

Pavement Storage

Infiltration Rate

In Situ Soil Compaction Associated with Lot Construction

• Initial In Situ Soil

Infiltration is INVARIABLY reduced when permeable pavements (&

• ther practices)

are installed.

www.bae.ncsu.edu/stormwat er

Tyner et al. (2009): Trenching the Subsoil

www.bae.ncsu.edu/stormwater

“Ripping” the Subsoil - Tyner et al. (2009)

www.bae.ncsu.edu/stormwater

www.bae.ncsu.edu/stormwater www.bae.ncsu.edu/stormwater

Permeable Pavement Design IWS - Shallow IWS - Deep Conventional

www.bae.ncsu.edu/stormwater

100 200 300 400 500 600 700 800 900 1000 CONVENTIONAL DEEP IWS SHALLOW IWS Milimeters Total Rainfall Total Outflow from Underdrains

8 Outflow Events

Hydrology

www.bae.ncsu.edu/stormwater

77% Reduction 100% Reduction 99.5% Reduction 1 Outflow Event 0 Outflow Events

Same is True with Bioretention

• Construction impact
• n bottom layer

www.bae.ncsu.edu/stormwater

How are BRC’s constructed?

• Bioretention

– Backhoe w/ arm – Scoop out soil & replace w/ sandy media

www.bae.ncsu.edu/stormwat er

Construction Impacts

• Bioretention

– Bucket contact with bottom layer

• Compaction &

Smearing using conventional “scoop” technique  decreased infiltration rates

www.bae.ncsu.edu/stormwat er

Innovative Construction Technique

• Less intensive bucket

contact when excavating final 1 ft

• f subsoil
•  “Rake” method

– Scarify surface with teeth on bucket

www.bae.ncsu.edu/stormwater

Tested Excavation Technique

• Scoop vs. Rake

– For final 9-12” of excavation, depth most affected by compaction

www.bae.ncsu.edu/stormwater

Varying… Soil Type

• Clay vs. Sand
• 2 Sites

1)Piedmont (clay): Raleigh 2)Coastal Plain (sand): Nashville

www.bae.ncsu.edu/stormwat er

And…Antecedent Moisture Condition

• Dry vs. Wet
• Effects of excavating right after a large rain or

if it rains before cell is filled

www.bae.ncsu.edu/stormwater

Compaction

• Scout SC-900 soil cone penetrometer
• < 1 minute per test

www.bae.ncsu.edu/stormwater

Nashville (Sand) “Wet” Cell

www.bae.ncsu.edu/stormwater

Infiltration

• Double Ring Infiltrometer

– Prevents divergent flow in middle ring  record rate of infiltration from inner ring

• ~ 90 minutes per test

www.bae.ncsu.edu/stormwater

Average Infiltration

www.bae.ncsu.edu/stormwater

• Sat. Hydraulic Conductivity
• Collect soil cores and run constant head

saturated conductivity test

www.bae.ncsu.edu/stormwater

Nashville Results

Site Type KSat Infiltration Bulk Density "Dry" / (Loamy-Sand) Performance Increase – Rake 84% 123%

• 2.4%

"Wet" / (sand) Performance Increase – Rake 172% 42%

• 3.8%

Notes: n = 6 n = 3 n = 6

• Average change in performance using rake
• vs. scoop method

www.bae.ncsu.edu/stormwater

Take Home Point

• How you Treat your soil

during construction makes a big difference in how well your SCM / LID Site is going to Perform

www.bae.ncsu.edu/stormwater

“Be nice to your soils!!!”

Summary

• Soils are an Integral Component of Low

Impact Development

– Establishing the Reference Hydrologic Condition – Set the Infiltration Capacity of SCMs – Influence the Pollutant Removal Ability

• Fill Soils/ Media Selection is critical for the

capture of several pollutants

• We can do a lot to “make soils” work for us.

And to overcome negative impacts of construction

www.bae.ncsu.edu/stormwater

Parting Thought: Know Soils… Know LID No Soils… No LID

www.bae.ncsu.edu/stormwater

Questions? Please Ask!